PROJECT SUMMARY Animals have numerous opportunities to acquire information from one another. Although common across taxa, social learning requires the integration of associative learning, memory and social behavior (LMS); and these vary considerably among individuals within a species. How do genetic differences cause alterations in neural functions that ultimately cause variability in social learning? To address this question it is important to identify the genes and neural circuits that play a role in LMS. Here we propose to analyze the molecular-genetic basis that underlies individual differences in LMS in the fruit fly Drosophila melanogaster. Flies have been used extensively as a model to study the genetic basis of development, behavior and learning and flies are social animals that aggregate in large groups. This confluence provides an opportunity to examine how individual genetic differences underlie variation in sociality and learning. We will develop flies as a model for the integrated analyses of the genetic, biochemical, physiological, and environmental components of LMS. We hypothesize that transcription in specific neural circuits is modified by experience and this contributes to long- term memory. Further, we hypothesize that variation in transcription in these same neural circuits across individuals contributes to the differences in LMS. In Drosophila, the mushroom body (MB) is the neural center of learning and cognition. Using a new molecular-genetic technique that allows for the purification of RNA from subsets of cells, we will examine the variation in both the MB transcriptome and the LMS behaviors of 192 recurrent heterozygous F1 genotypes. These F1 individuals will be obtained from progeny of 192 sequenced natural Raleigh strains, each crossed to a sequenced w1118 strain. We will be able to: i) determine how transcriptional variation in LMS is associated with phenotypic variation, ii) partition the cis- and trans- components of expression variation for every gene, iii) focus on cis- expression variation that is associated with behavioral variation, and iv) pinpoint cis- DNA polymorphisms likely contributing to cis- expression variation. Once we have identified causes of MB expression variation, we will connect the transcriptome with LMS: our goal is to identify the candidate genes and their polymorphisms associated with differences in LMS. We will measure larval and adult behaviors of focal genotypes with natural and major-effect alleles in the LMS candidate genes in several social contexts. We will then determine the transcriptional differences in the MB that are due to these social encounters. Subsequent functional analyses of candidate genes will provide mechanistic knowledge about how genes contribute to LMS. In sum, we propose an investigation where we synthesize genetic, molecular and behavioral information about individual genotypes to ultimately decipher aspects of the population genetics of social learning. We believe our inferences will illuminate, at several levels, the range of LMS variation maintained in natural populations of flies, thus providing insight into the genetic basis of LMS variation in other social organisms, including humans.